Welcome to the November round-up! There is a real wealth of interesting articles this month – these are my pick but do let us know what you have found useful.
Intervention – The big news this month was the reporting of the DAWN trial in the New England Journal of Medicine. The results had been eagerly awaited as it was strongly suspected that the timeframe for effectiveness of mechanical thrombectomy is longer than the traditional 6 hours, and sure enough there was significantly increased functional independence and reduced disability in the group that had a thrombectomy at 6 to 24 hours compared to the control group, with no significant difference in mortality or symptomatic haemorrhage. Importantly, the inclusion criteria specified demonstration of a mismatch between clinical severity and infarct volume as measured using CT perfusion or MRI DWI. This will obviously have major implications for organisation of stroke imaging and thrombectomy services.
Supportive evidence was also provided this month in the form of a subgroup analysis of the ESCAPE trial published in AJNR, which demonstrated a benefit for treatment within an extended time window (5.5-12 hours).
Vascular – Predicting progression of infarct is an important aspect of treatment and an article by Bouluois et al. in JAMA Neurol addresses a crucial issue, that if patients are transferred their infarct may progress beyond potential for treatment during the transfer. They assessed imaging factors associated with this progression in transferred patients, and found that collateral supply and clinical severity were key factors in determinants of infarct progression, suggesting a role for collateral assessment (they used a modified version of the Maas rating scale) in baseline stroke imaging.
Arnoux et al. in Stroke examined the imaging characteristics of the brain parenchyma related to cerebral venous thrombosis, looking at their own case series and identifying useful imaging features. They found that haemorrhagic ischaemia was the most common associated feature (followed by intracranial haemorrhage and non-haemorrhagic ischaemia), and determined six foci where parenchymal abnormalities occurred, with the inferior parietal lobule being the most common, and correlation between vessel and anatomical area.
Emergency – As everyone at RSNA this year will be acutely aware, artificial intelligence and deep learning are here to stay and an interesting article in Radiology by Prevedello et al. provides one example of how it may be used. They developed a system for identifying major abnormalities and showed good sensitivity and specificity for haemorrhage, mass effect and hydrocephalus (90% and 85% respectively) and reduced sensitivity but increased specificity for acute infarct (62% and 96%). This was for thick-section non contrast CT head, and the authors suggest this could help speed up workflow by highlighting critical cases to the radiologist.
Also in Radiology, Rodriguez-Luna et al. assessed the utility of the ‘spot sign’ to predict expansion of intracranial haemorrhage, and found that the use of multiphase CT angiography allowed greater stratification of risk with respect to the spot sign, with a more arterial ‘spot’ more predictive of haemorrhagic expansion.
Tumour – There is plenty to choose from this month, starting with a focused review on the aspects of the 2016 WHO classification of brain tumours relevant to radiologists in Radiographics. This is complemented by a more detailed review in Neurographics on the molecular subgroups of medulloblastoma described in the 2016 classification, covering specific imaging characteristics.
Also in Neurographics, Shin et al. provide an excellent overview of CNS lymphoma and its manifold presentations in immunocompetent and immunocompromised patients, with reference to advanced imaging techniques.
Finally, the humble vestibular schwannoma gets a whole review to itself in AJNR, covering diagnosis and imaging pathway but focusing on treatment options and follow-up, and is neatly accompanied by an article in JNS by Wu et al. specifically addressing post-Gamma Knife imaging appearances and predictors of treatment outcome.
Degenerative – The advent of susceptibility weighted imaging has led to an increase in the diagnosis of superficial siderosis, and an interesting study by Pichler et al. in Stroke investigated the prevalence in an elderly population. They found the incidence increased with age (0.21% at 50-69, 1.49% >69) and correlated with higher Apolipoprotein E e2 genotype and a higher likelihood of a positive amyloid PET, implicating cerebral amyloid angiopathy as the most common mechanism, and they therefore suggest follow-up imaging to assess risk of further haemorrhage.
The many different causes of dementia are covered in a comprehensive review by Desai et al. in Neurographics – the detailed table covering the imaging and clinical features of the different aetiologies is a particularly handy quick reference guide.
Paediatrics – Adams et al. present an in-depth review of congenital brain stem abnormalities, and manage to make a complex topic admirably clear and practical, using the embryological and molecular underpinnings of the brain stem to classify the imaging phenotypes.
In Pediatric Radiology, Linscott et al. examine venous disease in children, covering the pathophysiology and the imaging features of acute and chronic venous injury.
Miscellaneous – An MDT can be a great opportunity to learn and share knowledge with fellow clinicians, or can be an opportunity for humiliation for an unsuspecting trainee. A survey by the JRF published in the BJR looks at the trainee experience of MDTs and subsequently provides a guide to approaching MDTs based on their findings. This can often be insufficiently covered in training and the article provides a good opportunity to think about training needs in this context, and is well worth a read.
That’s all for now, see you next month!